Skip to main content

Full text of "Island rodents: a new species of Octodon from Isla Mocha, Chile (Mammalia: Octodontidae)"

See other formats

Z. Saugetierkunde 59 (1994) 27-41 
© 1994 Verlag Paul Parey, Hamburg und Berlin 
ISSN 0044-3468 

Island rodents: a new species of Octodon from Isla Mocha, Chile 
(Mammalia: Octodontidae) 

By R. Hutterer 

Zoologisches Forschungsinstitut und Museum Alexander Koenig, Bonn, Germany 

Receipt of Ms. 10. 2. 1993 
Acceptance of Ms. 22. 3. 1993 


A hitherto unknown species of Octodon occurs on Isla Mocha, a small coastal island in the Valdivian 
rainforest zone of central Chile. The new Pacific degu {Octodon pacificus n. sp.) exhibits characters 
such as soft and long fur, long and poorly tufted tail, homodont upper dentition, barely reduced third 
lower molars, broad and asymmetrical teeth with long reentrant folds full of cement, all of which are 
considered as plesiomorphic for the genus. It is suggested that geographic isolation led to the 
preservation of primitive characters. The new species probably represents the sister taxon of Octodon 
bridgesii, one of the three mainland species currently known from Chile and Argentina. The 
phylogenetic significance of the new degu is discussed. 


Hystricognath rodents of the family Octodontidae occur in west-central South America, 
from where six genera (Aconaemys, Octomys, Octodon, Octodontomys, Spalacopus, Tym- 
panoctomys) with ten species are known from Bolivia, Chile and Argentina (Pearson 
1984; Contreras et al. 1987; Mares and Ojeda 1982; Gallardo and Reise 1992). The 
systematic status and the contents of the family are rather controversial as no striking 
synapomorphies are known to characterize the group (Glanz and Anderson 1990); some 
authors include the Ctenomyidae (Reig 1970, 1986; Reig and Quintana 1991), others the 
Abrocomidae (Ellerman 1940). 

Confusion also exists at the species level. Gallardo and Reise (1992) have recently 
demonstrated that the genus Aconaemys comprises three species, not one or two, as 
previously thought. Three species are generally accepted in the genus Octodon: O. degus 
(Molina, 1782), O. bridgesii (Waterhouse, 1844), and O. lunatus Osgood, 1943 (Osgood 
1943; Mares and Ojeda 1982; Patterson and Feigl 1987; Reise and Venegas S. 1987; 
Gallardo 1992; Redford and Eisenberg 1992). Contreras et al. (1987) questioned the 
validity of O. lunatus on morphological grounds but the form has a distinctive karyotype 
(Spotorno et al. 1988) and certainly is a valid species. However, our knowledge on the 
taxonomy, distribution and ecology of this genus is far from being complete, as this report 
will show. New material already collected in 1959 from a small island off central Chile 
demonstrates the existence of a fourth species of Octodon. Its characters considerably 
enlarge the morphological diversity of the genus. The new degu is named and described 
below and its significance discussed in the context of the family. 

Material and methods 

The specimens studied are stored in the collections of the Museum Alexander Koenig, Bonn (ZFMK), 
the Staatliches Museum fiir Naturkunde Stuttgart (SMNS), the Senckenberg-Museum Frankfurt 
(SMF), and the collection of W. von Koenigswald, Paleontological Institute, University of Bonn 
(VKB). The following specimens were used for illustrations and for comparison: Octodon degus, 9 

U.S. Copyright Clearance Center Code Statement: 0044 - 3468 / 94 / 5901-0027 $ 02 . 50/0 


R. H utter er 

(ZFMK 86.89, 86.701-3, 87.82, 87.361, 88.37; SMNS 1113, 33771), O. bridgesii, 9 (ZFMK 88.63, 
88.83, 92.309, 92.382; SMNS 534; SMF 11037; VKB 972), O. lunatus, 2 (SMNS 42960; SMF 862), 
O. sp., 4 (ZFMK 92.383-6), Aconaemys fuscus, 1 (ZFMK 88.59), Spalacopus cyaneus, 1 (ZFMK 
92.310). Other sources of information were the descriptions and figures in Thomas (1920), Ellerman 
(1940), Wood (1949, 1974), Walker et al. (1964), Landry (1970), Reig (1970), Woods and 
Boraker (1975), Glanz and Anderson (1990), Nowak (1991), Reig and Quintana (1991), and de 
Santis et al. (1991). The terminology of the skull and the teeth follows Wood and Wilson (1936) and 
Woods and FIowland (1979). All measurements are in millimetres. A note is required about the use 
of the name Aconaemys in this report. According to Reig (1986) the genus Pithanotomys, based on a 
fossil, has priority over Aconaemys. Ffowever, evidence which the author announced to be presented 
in a forthcoming paper was never published except for the same statement plus one figure of upper and 
lower dentition in Reig and Quintana (1991). Although the similarity between the molars of the 
extant Aconaemys fuscus and the Pliocene Pithanotomys columnaris is striking, the shape and structure 
of the skull has not been described. I therefore refrain from following Reig and Quintana (1991) 
until a thorough comparison of both extant and fossil forms has been published. 

Results and discussion 

Octodon padficus, new species 

Holotype: ZFMK 92.384, skin and skull of an adult female, collected by Francisco Behn 
on 16 January 1959, field number L 6. The skin is in good condition, the cranium (Fig. 2) 
lacks the occipital and the bullae; the mandible is complete. 

Paratypes: Skins and skulls of another adult female (ZFMK 92.383; Fig. 1) and of two 
juveniles (ZFMK 92.385-6), collected between 11 and 24 January 1959 by F. Behn. 

Measurements: See tables 1 to 3. 

Type locality: Isla Mocha (38°22' S, 73°55' W), Arauca Province, Chile. Mocha Island is 
situated 31 km off the coast; its maximum extension from north to south is 13 km, and 
5-7 km from west to east. The centre of the island forms a plateau of about 20 square 
kilometres which is almost entirely covered by myrtle forest (Mrs. Erika Behn, in litt.). 
Two peaks ascend to an altitude of 323 m in the north and 390 m in the south of the island. 
A lake is on top of the hills. Valleys run down from these peaks to the eastern coast. Large 
meadows cover the coastal plains of the island. Dr. Behn did not specify where he 
collected the small mammals but a recent map shows two airstrips on the eastern side of the 
island and one of them may have been the meadow where his expedition landed and 
camped. Bullock (1935) described the island as a hilly plateau covered with virgin forest, 
almost inacessable except where the inhabitants cut tracks into the forest. A short 
description of the island and its vegetation is also provided by Almeyda Arroyo (1955). 
Translated from Spanish it reads: “To the eye of the seaman Mocha Island presents a 
beautiful green aspect: the hills up to the peaks and the slopes towards the sea are covered 
with large trees, providing easy access to wood.” Isla Mocha is near the northern limit of 
the Valdivian rainforest zone, as outlined by Osgood (1943). Rocks and sediments of the 
island are of Miocene, Pliocene and Quaternary ages (Tavera and Veyl 1958). 

Diagnosis: Larger and heavier than the three other species of Octodon; fur uniformly dark 
brown washed with orange, hairs soft and long; tail long (77 % of head and body length) 
with inconspicuous tuft. Skull long, particularly the diastema; zygomatic arches wide in 
dorsal and straight in lateral view; zygomatic process of squamosal inserting very high; 
superior jugal process forming a characteristic spine. Upper cheek-teeth very broad and 
uniform, the long reentrant folds filled with cement; lower cheek-teeth more or less 
homodont, their folds running strongly oblique and almost parallel; reentrant folds also 
cemented; the third lower molar is large and has the shape of an Arabic numeral 1. 

Description: Although Octodon pacificus n. sp. is larger and heavier than O. bridgesii, O. 
lunatHS and O. degus (Tab. 1), its body appears to be more slender (Fig. 1). The overall 

Island rodents: a new species of Octo don from Isla Mocha, Chile 







colour of the pelage is brown- 
orange, with the orange tips 
being brighter on the hairs of 
the underside. The dorsal 
hairs are 22 mm long on aver- 
age and very soft, scattered 
long guard hairs being up to 
27 mm. The basal 80 % of the 
dorsal hairs are plumbous, 
their tips are brown-orange. 

The head of the animal is uni- 
formly coloured; no pale eye 
rings are present. The rela- 
tively short dark ears appear 
almost naked, bearing only 
very fine short hairs. No 
white or yellow ear tufts exist. 

The snout bears about 30 vib- 
rissae, some of which are 
black and others white. The 
forefeet have 4 digits with 
claws and a tiny pollex with a 
very reduced nail. Their dor- 
sal surfaces are covered with 
greyish-brown hairs; they 
turn to white at the outer edge 
of each forefoot. The ventral 
surfaces are naked and show a 
granulation which is typical 
for the genus. Each palmar 
surface has three interdigital 
and two palmar pads. The 
hairs on the dorsal surfaces of 
the hindfeet are creamy- 
brown. A group of long and 
stiff white hairs sit on top of 
each digit. Finely granulated 
skin covers the space between 
the five interdigital and two 
plantar pads on the ventral 
surfaces of the hindfeet. The 
tail is long (77 % of head and 
body length), thinly haired, 
with the terminal 50 mm 
bearing a tuft of slightly 
longer and darker hairs. The 
more proximal part of the tail 
is brown on the dorsal and 

cream-brown on the ventral surface; the inconspicuous terminal tuft appears dark brown. 

The cranium of Octodon pacificus n. sp. is shown in Fig. 2. Unfortunately, the bullae 
and the occipital have not been preserved, due to the initial preparation of the animals in 
the style of a bird skin. The skulls were subsequently removed from inside the skins and 


Fig. 1. Octodon pacificus n. sp., dorsal and ventral aspect of the 
skin of paratype 2FMK 92.383; total length of the specimen is 
390 mm 


R. H utter er 

Fig. 2. Lateral view of a cranium of 
Octodon pacificus n.sp. (based on 
the holotype, missing parts of the 
shaded area reconstructed), and of a 
cranium of O. degus (below, 
adopted from Woods and Bora- 
KER 1975). Note the differences in 
size, diastema, infraorbital fora- 
men, and the position of the in- 
ferior rim of the zygomatic arch 


Fig. 3. A comparison of the zygomatic arch in the four species of Octodon. if: infraorbital foramen; 
ijp: inferior jugal process; j: jugal; 1: lacrymal; Ijf: lateral jugal fossa; sjp: superior jugal process; ssp: 
superior squamosal process; szr: superior zygomatic root; zpm: zygomatic process of maxillary; zps: 
zygomatic process of squamosal 

Island rodents: a new species of Octodon from Isla Mocha, Chile 


the skins reworked as standard 
study skins in 1992. The skull is 
large and broad, with a long 
diastema, a large infraorbital 
foramen, and a broad interorbital 
constriction. The lateral ridges of 
the frontals form a thin, translu- 
cent roof which extends further 
than in the other species. The 
measurements for the zygomatic 
width of both adult specimens 
exceed all known measurements 
for the other species (Tab. 2); the 
same holds for the length of the 
nasals. The tips of the nasals pro- 
ject further than in the other species of the genus (Fig. 2). In its general configuration the 
skull of Octodon pacificus n. sp. resembles more that of O. hridgesii. In side view the 
zygomatic arch is diagnostic: its jugal-maxillary part is very straight, as is the upper rim of 
the lateral jugal fossa; the insertion of the zygomatic process of the squamosal is very high, 
leaving considerable space between the inferior jugal process and the upper molar row 
(Figs. 2, 3). The superior jugal process forms a characteristic spine, less developed in the 
other species of the genus. The superior zygomatic root points slightly obliquely in 
anterior direction, whereas this bone is upright or even points posteriorly in the other 
species (Fig. 3). The superior squamosal process is poorly developed, much less so than in 
O. lunatus and O. degus. 

The upper incisors are thick and broad (Tab. 2) and their position is opisthodont; the 
anterior surfaces are stained orange, and the tips are deeply notched. The upper cheek- 
teeth are rootless (Fig. 4) and larger and especially broader than in the other species (Tab. 3, 
Figs. 5, 6). Both upper and lower teeth are more uniform in size and shape than in the 
other species (Fig. 4). In fact, the upper P4 to M3 show only minor differences in size, not 
in shape. All are highly asymmetrical with a heavy paracone and metacone and a long 
reentrant fold which is almost completely filled with cement (Fig. 6). The same is true for 
the lower cheek-teeth; here the two enamel folds run parallel to each other and strongly 
oblique. The third lower molar is somewhat simplified but keeps the size and aspect of the 

Table 1. External measurements and body mass of samples of Octodon; size data for species other 
than pacificus n. sp. taken from Redford and Eisenberg (1992), and body mass from Bozinovic 


O. pacificus 

n. sp. 

O. bridge sii 

O. lunatus 

O. degus 






Total length 









Tail length 









Hindfoot c.u. 


















Body mass (g) 










Ml P4 

Fig. 4. Isolated left upper and lower molars of Octodon 
pacificus n. sp. (left side) and O. hridgesii in lingual view 


R. H utter er 

Table 2. Cranial measurements of some specimens of Octodon spp. 

O. pacificus 

n. sp.^ 

O. hridgesiF 

O. lunatuF 

O. degus^ 

Greatest length 






Zygomatic width 






Interorbital width 



8.1- 9.0 



Nasalia length 






Nasalia width 



5.1- 5.8 



Diastema length 



8.1- 9.9 



Upper toothrow, crowns 






Upper toothrow, alv. 






Width P4-P4 



6.5- 7.7 



Width of both upper 11 



3.1- 3.7 



■ holo-, paratype. - ^ 6 skulls (ZFMK, SMNS, VKB). 

- ^ SMNS 42960. 

-‘‘ZFMK 86.701. 

Table 3. Width of upper (P4-M3) and lower (p4-m3) cheek-teeth of two adult specimens each of 
Octodon pacificus n. sp. (holo- and paratype), O. bridgesH^ and O. degus 

O. pacificus n. sp. 

O. bridgesii 

O. degus 


2.82, 2.85 

2.57, 2.50 

2.04, 2.11 


2.57, 2.70 

2.40, 2.29 

1.96, 1.98 


2.67, 2.62 

2.39, 2.31 

1.99, 1.96 


2.34, 2.28 

1.99, 1.84 

1.79, 1.66 


2.26, 2.22 

2.22, 2.30 

1.85, 1.89 


2.27, 2.47 

2.41, 2.24 

1.93, 2.15 


2.30, 2.45 

2.15, 2.16 

1.76, 1.62 


2.05, 2.33 

1.66, 1.74 

1.50, 1.56 

other molariform teeth. In occlusal view, this tooth resembles the Arabic numeral 1 

The mandible is larger and heavier than in the other species of the genus, as are the 
lower incisors. The condyloid process is particularly broad and heavy, corresponding to a 
large glenoid fossa of the squamosal, as may be inferred from the development of the 
posterior zygomatic root of the cranium (Fig. 3). 

Fig. 5. Occlusal view of upper and lower molars. Left: Octodon bridgesii (ZFMK 88.63), middle: O. 
pacificus n. sp. (ZFMK 92.384), right: O. pacificus n. sp. (ZFMK 92.383); for measurements, see 
table 2 

Island rodents: a new species of Octodon from Isla Mocha, Chile 


C D 

1 mm 

Fig. 6. Occlusal view of right upper and right lower molars of the four species of Octodon. A: O. 
pacificus n. sp. (holotype), B: O. pacificus n. sp. (paratype), C; O. bridgesii (ZFMK 88.63), D; O. 
lunatus (SMNS 42960), E: O. degus (ZFMK 86,701). The shading indicates cement 

Notes on juveniles: As mentioned above, two juveniles were collected together with the 
two adult females. Their general appearance matches that of the adults, except the fur is 
duller. Especially the venter is more grey and not washed with orange. The weights of the 
young were 50 and 55 g, respectively, and the upper and lower third molar were not 
erupted. Their age may be estimated at two or three weeks. Assuming this age, they would 
have been born around the last week of December. Taking a gestation time of 90 days as in 
Octodon degus (Weir 1974), conception would have occurred in September of the 
previous year. This coincides with the reproduction period of Octodon degus near 
Santiago, Chile (Fulk 1976). 


R. H utter er 

Comparisons: The three species of Octodon currently recognized can be distinguished 
from O. pacificHS n. sp. as follows: 

Octodon deguSy type species of the genus, is smaller and has a stouter body; it has a 
shorter tail with a thick terminal brush, a coarse, agouti-coloured fur, a lighter venter and 
white ear tufts. In five specimens examined the pollex bears a small but clearly pointed 
claw, which contrasts to the reduced nail of the new species. The statement of Osgood 
(1943) that the species of the genus Octodon bear a nail on the pollex is not always correct. 
Bennett (1832), when diagnosing the genus Octodon and his new species cumingii (a 
synonym of degus), correctly stated that “the thumb of the fore feet has a short obtuse 
claw”, not a nail. Nevertheless, Osgood’s statement is equally found in standard 
references such as Woods and Boraker (1975). It may be possible that the character is 
variable but this has to be checked in a larger series. The observed differences between O. 
degus (claw) and O. pacificus n. sp. (nail), however, are very clear and rather suggest the 
existence of a good diagnostic character. The skull of O. degus is smaller and the position 
of the zygomatic arch is lower (Fig. 2). This species has the simplest molars of the genus 
(Fig. 6): upper and lower molariform teeth are fairly symmetrical and approach an 8-shape 
more than any other Octodon. 

Octodon lunatus is similar to O. degus externally and cranially, although Osgood 
(1943) claimed that his new species was indistinguishable from O. hridgesii. The specimen 
at hand matches perfectly the description of Osgood, particularly with respect to the 
lunariform third upper molar (Fig. 6). Externally, the specimen differs from degus by a 
somewhat softer and longer pelage although the agouti colour is very similar. The 
underside of the skin, however, is paler and washed with cream-white. Yellow ear tufts are 
present. The dorsal surfaces of hands and feet are white. The pollex of the manus bears a 
nail, not a claw. The tail is long with a prominent terminal brush, dorsally dark brown, 
ventrally white on the first half and blackish-brown on the terminal part. Long whitish 
guard hairs are present mainly on the posterior part of the fur. The skull is similar to O. 
degus and O. hridgesii, but the M3 is diagnostic (Fig. 6). 

Octodon hridgesii is similar to O. pacificus n. sp. in the softer texture of the fur, and in 
its trend towards asymmetry of the teeth (Figs. 5, 6). In addition, the pelage is more 
uniform and lacks white or yellow ear tufts. The colour of the body hairs is a mixture of 
brown and yellow, not orange. The tail is shorter (75 % of head and body length) but is 
similar lo pacificus n. sp. in not having such a prominent terminal tuft (Waterhouse 1844; 
Mann 1958) as in O. degus and O. lunatus. Averages for Octodon hridgesii are smaller for 
all measurements (Tab. 2). Clear-cut differences to O. pacificus n. sp. exist in the size and 
shape of the skull, the shape and position of the zygomatic arch (Fig. 3), and in the size and 
form of the molariform teeth (Fig. 6), particularly in the upper and lower third molars. 

History of discovery: The discovery of this new rodent has been a matter of enthusiastic 
research effort and personal tragedy. One person involved is Dr. Francisco Behn (11. 6. 
1910-28. 5. 1976), formerly a Professor of Anatomy and Pathology at the University of 
Concepcion, Chile, and a free-time ornithologist. Although occupied by his profession he 
also spent his free time with the study of birds and such contributed to the ornithology of 
Chile (Bern and Millie 1957). Together with his wife Erika Bern he brought together 
an important collection of birds from northern Chile to Antarctica. For many years he 
corresponded with Dr. Gunther Niethammer (28. 9. 1908-14. 1. 1974), at that time 
curator of birds at the Museum Alexander Koenig, Bonn, whom he knew personally from 
a visit to Bonn in 1954, as evidenced by documents in the archives of the Bonn Museum. In 
1959 Dr. Bern and his family made an expedition to Isla Mocha where they camped for 
two weeks. On the first of September, 1959, he wrote to G. Niethammer [translated from 
German]: “All of January I visited an ornithologically extremely interesting island, which 
is situated some kilometres south of Concepcion off the mainland: the so-called Isla 

Island rodents: a new species of Octodon from Isla Mocha, Chile 


Mocha. It has been studied only once before in more detail by a bird watcher, since access 
is only possible by aircraft which has to land on the beach or on one of the meadows.” 
In his letter Bern enthusiastically continued to talk about his observations on the 
nesting habits of a seabird, Puffinus creatopus, which he planned to write up and publish as 
soon as his busy professional life would allow. For unknown reasons he never did so and 
all his experience was lost with his death in 1976. It remained also unknown that on his 
expedition to Isla Mocha he had collected a small number of mammals which he sent to 
G. Niethammer’s then 22 year-old son, Jochen Niethammer, who later became 
Professor of Zoology at Bonn University. The specimens remained unstudied for more 
than thirty years in his private collection and came to light only recently when it was 
transferred to and curated at the Museum Alexander Koenig, a final consequence of his 
tragic car accident which happened on an excursion with students in July 1991. Severe 
injuries terminated the career of this well-known mammalogist, main editor of the 
‘‘Handbook of European Mammals” and author of numerous papers on Palaearctic and 
African mammals, but also the discoverer of the extinct giant rat of the Galapagos Islands 
(Niethammer 1964). 

Etymology: The species is currently known only from a small island in the Pacific Ocean, 
hence its specific epithet. 

Other mammals: Very little is known on the remaining mammal fauna of Isla Mocha. 
Philippi (1900) described a new mouse as Mus mochae, which Osgood (1943) assigned to 
Akodon olivaceus (Waterhouse, 1837). Osgood (1943) also reported on three other rodent 
species obtained by D.S. Bullock in 1932 on this island. Apart from the new Octodon 
and samples of Akodon longipilis castaneus Osgood, 1943 (2FMK 92.387-389), and 
Akodon olivaceus mochae (Philippi, 1900) (2FMK 92.390-393), the small collection 
obtained by F. Bern in 1959 contains a specimen of Rattus rattus Linnaeus, 1758 (ZFMK 
92.394), a species which has not been reported from this island before. Table 4 summarises 

Table 4. Mammals recorded from Isla Mocha, based on Philippi (1990), Osgood (1943), and the 

present report 








Octodon pacificus n. sp. 


Akodon longipilis 



Akodon olivaceus 




Geoxus valdivianus 


Oryzomys longicaudatus 


Rattus rattus 


the small mammal species so far known from Isla Mocha. The cricetine rodents listed are 
characteristic for the temperate Valdivian rain forest on the neighbouring mainland 
(Meserve et al. 1982). However, the material from Isla Mocha stored at the American 
Museum, the British Museum, and the Bonn Museum deserves careful study of its own. 
Some subspecies named for populations of Isla Mocha may in fact represent full species. 
This seems probable for Akodon longipilis castaneus, the skull of which differs markedly 
from what Reig (1987) figured as representing the typical mainland population of A. 


R. H utter er 

Relationships within the genus Octodon 

Looking at morphological characters, the four extant species of Octodon fall into two 
groups. One includes O. hridgesii and the new species, the other O, degus and O. lunatus. 
The first two species share the following characters: uniform colouration, soft pelage, 
short ears, long but inconspicuously tufted tail, asymmetrical teeth with a long reentrant 
fold. Photographs of live O. hridgesii and O. degus in Reise and Venegas S. (1987) neatly 
illustrate the external differences. O. degus and O. lunatus share the more vivid coloura- 
tion, hairs of agouti-type, light eye marks, larger and tufted ears, pronounced black tail 
tufts, less asymmetrical teeth, and highly reduced third molars. The author regards most of 
the characters of the first group as primitive and those of the second group as derived for 
the genus. 

The supposed polarities are based on an outgroup comparison with fossil Octodontidae 
(Wood 1949; Pascual 1967; Patterson and Wood 1982; Reig and Quintana 1991), 
Ctenomyidae (Pascual et al. 1965; Reig 1970; Verzi et al. 1991) and Echimyidae 
(Patterson and Pascual 1968; Lavocat 1976), particularly with the Oligocene Platypit- 
tamys hrachyodon Wood, 1949, which is often taken as an ancestor model for the living 
Octodontidae. This assumption is justified because fossils of Platypittamys share a 
characteristic enamel structure with the extant Octodontoidea (Martin 1992), but not 
with other Caviomorpha. It should, however, be noted that Reig and Quintana (1991) 
presented a different view of molar evolution in octodontids which will be discussed 

An ingroup comparison reveals that Octodon pacificus n. sp. assembles more plesio- 
morphic characters than the other three species. The high position of the zygomatic 
process of the squamosal is only found in O. pacificus n.sp. (Fig. 2); it is shared with 
Platypittamys (Wood 1949). Platypittamys and some other fossil octodontids have 
strongly asymmetrical and homodont molars (Wood 1949). Within genus Octodon, O. 
pacificus n. sp. approaches these conditions more than the other species (Figs. 6, 8). A 
reduction of the third molars, as in O. degus and O. lunatus, is certainly a derived feature 
and may be a general evolutionary trend in octodontoid rodents. 

In conclusion, Octodon pacificus n.sp. may be regarded as the most primitive species of 
the genus. Geographically isolated and in the absence of similar-sized competitors the 
species may have lived on Isla Mocha since the Miocene, the geological age of the island 
(Tavera and Veyl 1958). Some morphological changes, however, must have occurred, as 
the large infraorbital foramen and the backward position of the superior zygomatic root 
(Fig. 3) are derived characters. O. pacificus n. sp. and O. hridgesii most probably had a 
common ancestor; both share a similar morphology and possibly similar ecological 
requirements. They have the southernmost distributions of the genus (Fig. 7) and seem to 
be more restricted to forest (Greer 1968; Meserve et al. 1982) than O. degus and O. 
lunatus which are adapted to life in semiarid shrublands (Woods and Boraker 1975; 
Contreras et al. 1987; Meserve and Le Boulenge 1987; Bozinovic 1992). 

How does this interpretation fit with the available chromosomal data? O. degus and 
O. hridgesii have 58 chromosomes (Gallardo 1992), while Octodon lunatus has 78 
(Spotorno et al. 1988) (Tab. 5). Gallardo (1992) discussed in detail the polarities of the 
karyotypes concluding that 58 represents the plesiomorphic condition and the higher 
number of 78 a derived condition. This view, which is in contradiction to Spotorno et al. 
(1988), is in full congruence with the morphological conclusions made above. 

Comparisons with other Octodontidae 

It is a matter of curiousity that the type genus of the family, Octodon, does not show the 
character for which it was named (Fig. 8). 8-shaped teeth are found in Aconaemys (Mann 

Island rodents: a new species of Octodon from Isla Mocha^ Chile 


Fig. 7. Maps of Chile and Argentina with the approximate distributions of the four species of Octodon 
indicated; in part adopted from Redford and Eisenberg (1992), modified 

1958; Glanz and Quintana 1991), and Spalacopus (Reig 1970), while Octodon (Figs. 6, 
8) and Octodontomys (Glanz and Anderson 1990) have asymmetrical teeth; the latter 
genus has no reentrant folds at all. If we look again at Platypittamys (Fig. 8), we would 
have to take Octodon as the most primitive genus of Octodontidae, with Octodontomys 
perhaps as an offshoot of Octodon^ and Aconaemys^ Octomys, Tympanoctomys, and 
SpalacopHs as members of a more derived clade. This grouping coincides largely with 
Thomas (1920), who divided the then known genera in a group with “crescentic” and 
another with “8-shaped teeth”. Reig and Quintana (1991) argued for the contrary. They 
did not mention Platypittamys but stated that 8-shaped molars, as in Aconaemys, are 
primitive and asymmetrical ones, as in Octodon, are derived, a view obviously taken from 
the Miocene Pseudoplataeomys elongatus which has perfectly 8-shaped molars (Reig and 
Quintana 1991). They also described a new genus, Ahalosia, from the Pleistocene of 
Argentina, which in their interpretation may have been the ancestor of genus Octodon. 
While the present author can concur that Pseudoplataeomys may have been ancestral to the 
group of octodontids with 8-shaped molars (Aconaemys, Octomys, Tympanoctomys, 
Spalacopus), it does not follow that Octodon is derived from the Pleistocene Ahalosia. This 
genus has very simple molars resembling Octodontomys (Fig. 8) but it also has very long 
upper third molars which are quite unusual for the group. Also the skull of Ahalosia, as 
figured by Reig and Quintana (1991), is not similar to Octodon but instead recalls the 
skull of Aconaemys with its short nasals, broad interorbital region, parallel molar rows, 
and stout mandible. The phylogenetic position of Abalosia should thus be regarded as 
uncertain and its postulated relation to Octodon cannot be accepted. If we include the 
Oligocene Platypittamys in this comparison then again Octodon would group next to it 
(Fig. 8). The Pliocene Chasicomys octodontiforme Pascual, 1967 shows a somewhat 
intermediate morphology of the upper molars (Pascual 1967) and may be taken as support 
for the supposed direction of molar evolution. It can be supposed that different groups of 


R. Hutterer 


D E F G 

Fig. 8. Right upper molars of octodontid rodents. A: Platypittamys hrachyodon (copied from Wood 
1949, slightly simplified); B: Octodon pacificus n.sp. (paratype); C: Octodontomys gliroides (copied 
from Reig and Quintana 1991, reversed); D: Spalacopus cyaneus (ZFMK 92.310); E: Tympanoc- 
tomys harrerae (copied from Reig and Quintana 1991, reversed); F: Aconaemys fuscus (ZFMK 
88.59); G: Octomys mimax (copied from Ellerman 1940, modified). Not to scale, tooth rows 
brought to approximately the same length for the purpose of comparison 

octodontids already diverged in the Miocene, and that their phylogeny is at present 
obscured by an incomplete fossil record. Verzi et al. (1991) also concluded from their 
work on fossil Octodontoidea that octodontine and ctenomyine rodents radiated contem- 
poraneously in the Miocene. 

If we neglect the molars and look only at the skull, then Tympanoctomys (Landry 
1957; DE Santis 1991) would be the candidate for the most primitive genus because of its 
small infraorbital foramen and the extremely anterior position of the superior zygomatic 
root, which is shared by Platypittamys but 
not by any of the other five genera. On the 
other hand, Tympanoctomys and Octomys 
(Thomas 1920; Walker et al. 1964) both 
have hypertrophied auditory bullae, cer- 
tainly derived structures related to their 
gerbil-like habitus. 

The chromosome complements, which 
surprisingly are known for all species ex- 
cept Octodon pacificus n.sp., present a 
slightly different picture (Tab. 5). 

AconaemySy Octomys, Spalacopus, and Oc- 
todon have very similar karyotypes, whilst 
Octodontomys has a lower (38) and Tym- 
panoctomys a much higher (102) chromo- 
some number. The latter two genera may 
be taken as derived in this character. 

None of the different data sets provides 
a convincing solution of the phylogenetic relations between the six extant genera of 
Octodontidae, nor does the work based on blood protein similarities of only three genera 
(Woods 1982). At present it seems impossible to solve the phylogenetic relationships 
within the Octodontidae with the given information. One reason may be that the six extant 
genera represent only a small portion of the past diversity, a view held also by Gallardo 
(1992) and suggested by the fossils which are already known (Mones 1986; Reig and 
Quintana 1991) although many more fossils may be expected. Nevertheless, study of the 
fossil record together with the extant species allows the formulation of hypotheses on 

Table 5. Chromosome numbers of the species 
of Octodontidae; adopted from Gallardo 
(1992), Gallardo and Reise (1992), and refe- 
rences cited therein 




Aconaemys fuscus 



Aconaemys porteri 



Aconaemys sagei 



Octomys mimax 



Spalacopus cyaneus 



Octodon degus 



Octodon hridgesii 



Octodon lunatus 



Octodontomys gliroides 



Tympanoctomys harrerae 



Island rodents: a new species of Octodon from Isla Mocha, Chile 


evolutionary trends which could be tested in the future against more complete data sets 
based on other character complexes. 

A note on island rodents and conservation 

Since Osgood’s comprehensive work (1943), the mammal fauna of Chile is regarded as 
one of the best-studied in South America (Patterson and Feigl 1987). The latter authors 
predicted that “further additions to the faunal list are apt to be those species with highly 
restricted geographic ranges, especially those in remote areas”. This is perfectly true in the 
case of Isla Mocha, and counts also for two other rodents discovered on islands in southern 
Chile: Akodon markhami (Pine 1973) and Akodon hershkovitzi (Patterson et al. 1984). 
While those rodents were collected rather recently, the collection of the Pacific degu dates 
already from 1959. No attempts have been made since to check the actual status of the new 
species. It is not known whether it still forages in the meadows or forest of Mocha Island, 
nor, in the positive case, how large the population may be. What seems certain is that the 
natural range of the species covers only a few square kilometres and for this reason alone it 
must be regarded as vulnerable. Soft fur, orange-brown colour and the long tail suggest 
that this degu lives in the forest, thickets or swamps, which are restricted habitats on the 
island. In a report on the endangered mammals of Chile Miller et al. (1983) classified one 
of the mainland species, Octodon hridgesii, as vulnerable. They stated that the range of the 
species had been much reduced, presumably by increased cultivation of its valley habitat. If 
this is true for a rather widespread species, then such threats may apply even more to a 
small island population. The sample of Francisco Behn also shows (Tab. 4) that black 
rats were already present on Isla Mocha in 1959. Numerous examples show how fast 
endemic mammal faunas on islands are destroyed by human occupation, just to mention 
the Galapagos (Steadman and Ray 1982), the West Indies (Morgan and Woods 1986), or 
the Canary Islands (Boye et al. 1992). In the case of Mocha Island, a considerate field 
survey is suggested to test whether the Pacific degu still exists. Chilean mammalogists 
should feel encouraged to study this aspect* and propose conservation measures, if 


I owe gratitude to Mrs. Erika Behn, who sent me valuable information on Isla Mocha and even a 
short film taken during the 1959 expedition in which she participated. Dr. F. Dieterlen, Dr. G. 
Storch and Prof. Dr. W. von Koenigswald provided important specimens from their collections 
for study. Dr. K. Busse shared his knowledge of Chile and procured rare literature, and Mrs. H. von 
I ssENDORFF, Dr. G. Peters and Dr. G. Storch critically read and improved the manuscript. The 
photographs are the work of Mr. H. Meurer. I am grateful to all of them. 


Nagetiere auf Inseln: eine neue Octodon-Art von der Isla Mocha, Chile (Mammalia: Octodontidae) 

Auf der 31 km vor der chilenischen Kiiste gelegenen Insel Mocha lebt ein bisher unbekannter Degu, 
der unter dem Namen Octodon pacificus n. sp. beschrieben wird. Es handelt sich um eine grofie, 
weichhaarige und langschwanzige Art von einheitlich braun-orangener Farbung, der helle Uberaugen- 
streifen oder Ohrbiischel fehlen. Weiterhin bemerkenswert sind die homodonte Backenbezahnung 
und die asymmetrische Form der molariformen Zahne, deren linguale Schmelzfalten vollstandig mit 
Zement ausgefiillt sind. Im Vergleich zu fossilen Octodontidae erweisen sich die meisten Merkmale 
der neuen Art als plesiomorph; nachstverwandt diirfte Octodon hridgesii sein, wahrend O. degus und 
O. lunatus einer anderen abgeleiteten Gruppe zugerechnet werden. Beziehungen der 6 rezenten 
Gattungen der Familie untereinander und zu fossilen Vertretern werden diskutiert. Der neue Pazifik- 
Degu ist offenbar eine Reliktart, die auf der kleinen Insel Mocha iiberlebt hat. Das vorhandene 
Belegmaterial wurde bereits 1959 von dem chilenischen Arzt Francisco Behn gesammelt. Der 
gegenwartige Status der Art ist nicht bekannt. 


R. Hutterer 


Almeyda Arroyo, E. (1955): Geografia de Chile. 16. ed. Santiago de Chile: Impr. Talleres. 

Behn, F.; Millie, G. R. (1957): Notes. In: Suplemento de Las Aves de Chile. Ed. by J. D. Goodall, 
A. W. Johnson, and R. A. Philippi B. Buenos Aires, Argentina. 

Bennett, E. T. (1832): Characters of a new genus of Rodent Mammalia, presented by Mr. Cuming. 
Proc. zool. Soc. London, 46-48. 

Boye, P.; Hutterer, R.; L6pez-MartInez, N.; Michaux, J. (1992): A reconstruction of the Lava 
mouse (Malpaisomys insularis), an extinct rodent of the Canary Islands. Z. Saugetierkunde 57, 

Bozinovic, F. (1992): Rate of basal metabolism of grazing rodents from different habitats. J. 
Mammalogy 73, 379-384. 

Bullock, D. S. (1935): Las aves de la isla de la Mocha. Rev. Chil. Hist. Nat. 39, 232-253. 

Contreras, L. C.; Torres-Mura, J. C.; Yanez, J.-L. (1987): Biogeography of octodontid rodents: 
an eco-evolutionary hypothesis. Fieldiana Zoology, N. S. 39, 401-411. 

Ellerman, j. R. (1940): The families and genera of living rodents. Vol. I. Rodents other than 
Muridae. London: The Trustees of the British Museum. 

Fulk, G. W. (1976): Notes on the activity, reproduction and social behavior of Octodon degus. J. 
Mammalogy 57, 495-505. 

Gallardo, M. H. (1992): Karyotypic evolution in octodontid rodents based on C-band analysis. J. 
Mammalogy 73, 89-98. 

Gallardo, M. H.; Reise, D. (1992): Systematics of Aconaemys (Rodentia, Octodontidae). J. 
Mammalogy 73, 779-788. 

Glanz, W. E.; Anderson, S. (1990): Notes on Bolivian mammals. 7. A new species of Ahrocoma 
(Rodentia) and relationships of the Abrocomidae. Amer. Mus. Novit. 2991, 1-32. 

Greer, J. K. (1968): Mamiferos de la Provincia de Malleco. Publ. Mus. “Dillman S. Bullock”, Angol 
12, 1-114. 

Landry, S. O. (1957): The relationship of Petromys to the Octodontidae. J. Mammalogy 38, 351-361. 

Lavocat, R. (1976): Rongeurs caviomorphes de POligocene de Bolivie II. Rongeurs du Bassin 
Deseadien de Salla-Luribay (Collection Hoffstetter). Palaeovertebrata 7, 15-90. 

Mann, F. G. (1958): Clave de determinaclon para las especies de Mamiferos Silvestres de Chile. Trab. 
Centro Invest. Zool. Univ. Chile 3, 1-38. 

Mares, M. A.; Ojeda, R. A. (1982): Patterns of diversity and adaptation in South American 
histricognath rodents. Special Publ. Ser. Pymatuning Lab. Ecol., Univ. Pittsburgh 6, 393-432. 

Martin, T. (1992): Schmelzmikrostruktur in den Inzisiven alt- und neuweltlicher hystricognather 
Nagetiere. Palaeovertebrata, Mem. extra. 1-168. 

Meserve, P. L.; Le Boulenge, E. (1987): Population dynamics and ecology of small mammals in the 
northern Chilean semiarid region. Fieldiana Zoology, N. S. 39, 413-431. 

Meserve, P. L.; Murua, R.; Lopetegui, N. O.; Rau, J. R. (1982): Observations on the small 
mammal fauna of a primary temperate rain forest in southern Chile. J. Mammalogy 63, 315-317. 

Miller, S. D.; Rottmann, J.; Raedeke, K. J.; Taber, R. D. (1983): Endangered mammals of Chile: 
status and conservation. Biol. Conserv. 25, 335-352. 

Mones, a. (1986): Palaeovertebrata sudamericana. Catalogo sistematico de los vertebrados fosiles de 
America del Sur, parte I. Lista preliminar y bibliografia. Courier Forschungsinst. Senckenberg 82, 

Morgan, G. S.; Woods, C. A. (1986): Extinction and the zoogeography of West Indian land 
mammals. Zool. J. Linn. Soc. 28, 167-203. 

Niethammer, j. (1964): Contribution a la connaissance des mammiferes terrestres de Pile Indefatig- 
able (= Santa Cruz), Galapagos. Mammalia 30, 441-445. 

Nowak, R. M. (1991): Walker’s mammals of the world. 5th ed. Baltimore: The Johns Hopkins 
University Press. 

Osgood, W. H. (1943): The mammals of Chile. Field Mus. Nat. Hist., Zool. Ser. 30, 1-268. 

Pascual, R. (1967): Los roedores Octodontoidea (Caviomorpha) de la formacion Arroyo Chasico 
(Plioceno inferior) de la Provincia de Buenos Aires. Revista Mus. de la Plata (N.S., Paleontol.) 5, 

Pascual, R.; Pisano, J. A.; Ortega, E. J. (1965): Un nuevo Octodontidae (Rodentia, Caviomorpha) 
dp la formacion Epecuen (Plioceno medio) de Hidalgo (Provincia de La Pampa). Ameghiniana 4, 

Patterson, B. D.; Feigl, C. E. (1987): Faunal representation in museum collections of mammals: 
Osgood’s Mammals of Chile. Fieldiana, Zoology, N. S. 39, 485^96. 

Patterson, B. D.; Gallardo, M. H.; Freas, K. E. (1984): Systematics of mice of the subgenus 
Akodon (Rodentia: Cricetidae) in southern South America, with the description of a new species. 
Fieldiana, Zoology, N. S. 23, 1-16. 

Patterson, B.; Pascual, R. (1968): New echimyid rodents from the Oligocene of Patagonia, and a 
synopsis of the family. Breviora 301, 1-14. 

Island rodents: a new species of Octodon from Isla Mocha, Chile 


Patterson, B; Wood, A. E. (1982): Rodents from the Deseadan Oligocene of Bolivia and the 
relationships of the Caviomorpha. Bull. Mus. Comp. Zool. 149, 371-543. 

Pine, R. H. (1973): Una nueva especie de Akodon (Mammalia: Rodentia: Muridae) de la isla de 
Wellington, Magallanes, Chile. Anales Inst. Patagonia 4, 423-426. 

Philippi, R. A. (1900): Figuras i descripciones de los murideos de Chile. Anal. Mus. Nac. Chile, Ent. 
14a, Zool., 1-70. 

Redford, K. H.; Eisenberg, J. F. (1992): Mammals of the Neotropics. Vol. 2. The southern cone: 
Chile, Argentina, Uruguay, Paraguay. Chicago: The University Chicago Press. 

Reig, O. a. (1970): Ecological notes on the fossorial octodont rodent Spalacopus cyaneus (Molina). J. 
Mammalogy 51, 592-601. 

— (1986): Diversity patterns and differentiation of high Andean rodents. In: High altitude tropical 
biogeography. Ed. by F. Vuilleumier and M. Monasterio. New York: Oxford University 
Press. Pp. 404-439. 

— (1987): An assessment of the systematics and evolution of the Akodontini, with the description of 
new fossil species of Akodon (Cricetidae: Sigmodontinae). Fieldiana Zoology, N. S. 39, 347-399. 

Reig, O. A.; Quintana, C. A. (1991): A new genus of fossil octodontine rodent from the Early 
Pleistocene of Argentina. J. Mammalogy 72, 292-299. 

Reise, D.; Venegas S., W. (1987): Catalogue of records, localities and biotopes from research work 
on small mammals in Chile and Argentina. Gayana, Zool. 51, 103-130. 

Santis, L. J. M. de; Roig, V. G.; Justo, E. R. (1991): La anatomia craneo-dentaria de Tympanoc- 
tomys harrerae (Lawrence). Comparacion con Octomys mimax y consideraciones acerca de su 
est[u]d[i]o taxonomico (Rodentia: Octodontidae). Neotropica (La Plata) 37, 113-122. 

Spotorno, a.; Walker, L.; Contreras, L,; Pincheira, J.; Fernandez-Donoso, R. (1988): 
Chromosomas ancestrales en Octodontidae y Abrocomidae. Arch. Biol. Med. Exper. 21, 527. 

Steadman, D. W.; Ray, C. E. (1982): The relationships of Megaoryzomys curioi, an extinct cricetine 
rodent (Muroidea: Muridae) from the Galapagos Islands, Ecuador. Smiths. Contrib. Paleobiol. 
51, 1-23. 

Tavera, j.; Veyl, C. (1958): Reconocimiento geologico de la Isla Mocha. An. Fac. Cienc. Fis. Math., 
Univ. Chile 14-15, 157-188. 

Thomas, O. (1920): On mammals from near Tinogastra, Catamarca, collected by Mr. R. W. Hendee 
near Lake Junm. Ann. Mag. nat. Hist. (9) 17, 313-318. 

Verzi, D. H.; Montalvo, C. L; Vucetich, M. G. (1991): Nuevos restos de Xenodontomys simpsoni 
Kraglievich y la sistematica de los mas antiguos Ctenomyinae (Rodentia, Octodontidae). 
Ameghiniana 28, 325-331. 

Waterhouse, G. R. (1844): On the various skins of Mammalia from Chile, with notes relating to 
them by Mr. Bridges. Proc. zool. Soc. London 12,. 153-157. 

Walker, E. P.; Warnick, F.; Hamlet, S. E.; Lange, K. L; Davis, M. A.; Uible, H. E.; Wright, P. 
F. (1964): Mammals of the world, vol. 2. Baltimore: The Johns Hopkins Press. 

Weir, B. J. (1974): Reproductive characteristics of hystricomorph rodents. In: The biology of 
hystricomorph rodents. Ed. by I. W. Rowlands and B. J. Weir. Symp. Zool. Soc. London 34. 
London: Academic Press. Pp. 265-301. 

Wood, A. E. (1949): A new Oligocene rodent genus from Patagonia. Amer. Mus. Novit. 1453, 1-54. 

— (1974): The evolution of the Old World and New World hystricomorphs. In: The biology of 
hystricomorph rodents. Ed. by I. W. Rowlands and B. J. Weir. Symp. Zool. Soc. London 34. 
London: Academic Press. Pp. 21-60. 

Wood, A. E., Patterson, B. (1959): The rodents of the Deseadan Oligocene of Patagonia and the 
beginnings of South American rodent evolution. Bull. Mus. comp. Zool. 120, 279-428. 

Wood, A. E.; Wilson, R. W. (1936): A suggested nomenclature for the cusps of the cheek-teeth of 
rodents. J. Paleont. 10, 388-391. 

Woods, C. A. (1982): The history and classification of South American hystricognath rodents: 
reflections on the far away and long ago. Special Publ. Ser. Pymatuning Lab. EcoL, Univ. 
Pittsburgh 6, 377-492. 

Woods, C. A.; Boraker, D. K. (1975): Octodon degas. Mammalian Species 67, 1-5. 

Woods, C. A., Howland, E. B. (1979): Adaptive radiation of capromyid rodents: anatomy of the 
masticatory apparatus. J. Mammalogy 60, 95-116. 

Author's address: Dr. Rainer Hutterer, Zoologisches Forschungsinstitut und Museum Alexander 
Koenig, Adenauerallee 162, D-53113 Bonn 1, Germany